Automated production lines typically include many stations at which work is performed on a workpiece by a series of tools. A workpiece is clamped in position at each work station, and the tools at the several work stations engage the workpiece. The tool is then withdrawn at each work station, and the workpieces are unclamped and transferred simultaneously to the next work stations by a single transfer mechanism. The automotive industry is an example of an industry with high usage of automated production lines to produce engine blocks, for example. An engine block casting is passed successively through a series of work stations which mill, bore, drill, hone, gage, etc. It is not uncommon to have sixty work stations in such an automated production line. The times required to perform the several operations may be different, but the parts cannot be unclamped and transferred until all stations are finished because motion of a single transfer mechanism moves all parts.
Individual machine control systems generally include an automatic mode of operation wherein the machine is automatically cycled through a work sequence. Relay ladder logic commonly has been used to define the machine sequences, whether the logic is effected by relays or programmable controllers. The typical relay ladder logic diagram is a massive listing of relay, switch and solenoid conditions without an indication of logic flow which, while easy to read, is difficult to translate into the logic conditions intended to be represented. In addition, a large number of implied conditions exist that are not depicted. For example, a relay ladder with twenty elements (switches, relays, solenoids, etc.) embraces well over one million possible combinations of conditions. As a practical matter only a small fraction of these conditions, those necessary to make the system work, are considered by the machine designer and encoded into the logic scheme. Potential problems abound. Some planned conditions may be omitted, and unplanned conditions may be present, all with potentially serious consequences.
The programmable logic controller (PLC) was introduced with the hope that the massive racks of relays wired together in a permanent logic network could be replaced by a more reliable, smaller and readily reprogrammable electronic package. Although the PLC was designed to replace relay controls, its design explicitly seeks to replicate, in the design media, the relay ladder logic used by the technicians. Reprogrammability has proven to be a mixed blessing. Initial reprogramming is always needed as the system is set up and to meet product design changes. Any change made to the program logic requires a corresponding modification to associated diagnostic programming so the latter can properly reflect errors in the altered logic. The ease of making changes can allow modifications to be made by any semi-qualified person. Unfortunately, there is often no record of changes made, and there is therefore the possibility that the documentation may not reflect the actual program. This is a sufficiently serious problem for many production facilities that they are investing heavily in add-on equipment to monitor program changes. Furthermore, the increase in number of Input/Output (I/O) points desired by users and consequent greater complexity of controls have driven PLC manufacturers to push the PLC design to larger and larger units. These very large units are well designed to meet the needs of central chemical plant control rooms, for example, but are too slow for the fast-cycle automation, such as transfer lines.
A significant cost of automated production lies in system "down-time"--i.e. time in which the automated system is non-productive--which may range between 50% and 60% during production shifts. Total down-time may be visualized as consisting of three segments: diagnosis, repair and restart. Analyses show that there are many causes of down-time including: many potential sources of unintended stoppage, variable delays in bringing the appropriate skills to bear on the problem, repair times that vary from minutes to hours, and a significant restart time needed to return the machine to a "ready for automatic cycle" condition after all repairs have been made. Down-time is a direct consequence of the high degree of complexity required to perform a large number of machining and part handling operations with the minimum of operator intervention. While lower equipment failure rates are a high priority objective, it is widely believed that reducing time to return to production--i.e., restart time--is the major area of opportunity for improvement.
Thus, a particular problem which has plagued the art involves restart of an automated system after shutdown. Typically, operators or technicians must examine the status of workpieces at the individual stations and manually cycle the individual stations so that all stations are at a common point in the overall operating cycle. In a system which includes sixty work stations, for example, substantial time is required to restart the system in this manner after shutdown. Furthermore, in the event that an error is made and a workpiece is transferred to the next work station without the previous operation having been performed, substantial damage is likely to both the workpiece and the operating mechanism at the station, resulting not only in further shutdown of the system but in costly repair.